Hypoxia Signaling Cascade for Erythropoietin Production inHepatocytes

Yutaka Tojo,a,b,c Hiroki Sekine,a,b Ikuo Hirano,b Xiaoqing Pan,a,b* Tomokazu Souma,b* Tadayuki Tsujita,b,d Shin-ichi Kawaguchi,d*Norihiko Takeda,e Kotaro Takeda,f Guo-Hua Fong,f Takashi Dan,d Masakazu Ichinose,c Toshio Miyata,d Masayuki Yamamoto,b

Norio Suzukia

Division of Interdisciplinary Medical Science,a Department of Medical Biochemistry,b Department of Respiratory Medicine,c and Division of Molecular Medicine andTherapy,d Tohoku University Graduate School of Medicine, Sendai, Japan; Department of Cardiovascular Medicine, Graduate School of Medicine, University of Tokyo,Tokyo, Japane; Center for Vascular Biology, Department of Cell Biology, University of Connecticut Health Center, Farmington, Connecticut, USAf

Erythropoietin (Epo) is produced in the kidney and liver in a hypoxia-inducible manner via the activation of hypoxia-inducibletranscription factors (HIFs) to maintain oxygen homeostasis. Accelerating Epo production in hepatocytes is one plausible thera-peutic strategy for treating anemia caused by kidney diseases. To elucidate the regulatory mechanisms of hepatic Epo produc-tion, we analyzed mouse lines harboring liver-specific deletions of genes encoding HIF-prolyl-hydroxylase isoforms (PHD1,PHD2, and PHD3) that mediate the inactivation of HIF1� and HIF2� under normal oxygen conditions. The loss of all PHD iso-forms results in both polycythemia, which is caused by Epo overproduction, and fatty livers. We found that deleting any combi-nation of two PHD isoforms induces polycythemia without steatosis complications, whereas the deletion of a single isoform in-duces no apparent phenotype. Polycythemia is prevented by the loss of either HIF2� or the hepatocyte-specific Epo geneenhancer (EpoHE). Chromatin analyses show that the histones around EpoHE dissociate from the nucleosome structure afterHIF2� activation. HIF2� also induces the expression of HIF3�, which is involved in the attenuation of Epo production. Theseresults demonstrate that the total amount of PHD activity is more important than the specific function of each isoform for he-patic Epo expression regulated by a PHD-HIF2�-EpoHE cascade in vivo.

Under low-oxygen conditions (hypoxia), the number of circu-lating red blood cells increases to deliver oxygen efficiently

into peripheral organs. The red blood cell mass is controlled by theerythroid growth factor erythropoietin (Epo), the majority ofwhich is produced by the kidneys in a hypoxia-dependent manner(1). Therefore, inhabitants of high-altitude areas and patients suf-fering from chronic respiratory failure with chronic obstructivepulmonary disease often develop polycythemia (erythrocytosis)with elevated levels of Epo in their plasma (2, 3). Epo binds to itsreceptor (EpoR) on the surfaces of immature erythroid cells andstimulates signaling cascades for proliferation, differentiation,and antiapoptosis (4). Epo or EpoR gene-targeted mouse lines areembryonic lethal, with severe anemia around embryonic day 13,clearly indicating a requirement for Epo signaling in erythropoi-esis (5).

Anemia often occurs in patients who suffer from kidney dam-age (6). We previously reported that renal Epo-producing (REP)cells that are in the interstitial spaces between renal tubules trans-form into myofibroblastic cells, which are closely associated withrenal fibrosis under inflammatory conditions (7–11). Because thetransformed REP cells lose their ability to produce Epo even underseverely hypoxic conditions, pharmacologically inducing Epoproduction in the damaged kidneys of anemic patients is difficult.Although REP cells secrete most of the Epo in adult animals, hepa-tocytes are the primary Epo-producing cells in fetuses (12). Theliver maintains its Epo-producing activity throughout adulthood;thus, Epo expression is detectable in the livers of anemic/hypoxicmice (8, 12). However, the level of hepatic Epo production is weakand insufficient to compensate for renal anemia (13). Therefore,pharmacological enhancement of hepatic Epo production is a rea-sonable strategy for treating anemic patients who have renal dis-eases (14, 15).

In hepatocytes, Epo transcription, which is the rate-limitingstep of Epo production, is controlled by a cis-regulatory elementthat is proximally downstream of the Epo transcription end site(called Epo hepatic enhancer [EpoHE] in this study). EpoHE con-tains a binding sequence for hypoxia-inducible transcription fac-tors (HIFs) (16). We previously demonstrated that EpoHE is nec-essary and sufficient for Epo expression in hepatocytes, whereasrenal Epo production is independent of EpoHE (12, 13). Thus,while Epo expression is commonly induced by hypoxic stresses inREP cells and hepatocytes, these two cell types employ differentmechanisms of Epo gene regulation.

HIFs are master regulators of hypoxia-inducible gene expres-sion. Each HIF consists of an oxygen-responsive � subunit and aconstitutively expressed nuclear � subunit (17). Mammalian ge-

Received 11 February 2015 Returned for modification 16 March 2015Accepted 22 May 2015

Accepted manuscript posted online 26 May 2015

Citation Tojo Y, Sekine H, Hirano I, Pan X, Souma T, Tsujita T, Kawaguchi S-I,Takeda N, Takeda K, Fong G-H, Dan T, Ichinose M, Miyata T, Yamamoto M, SuzukiN. 2015. Hypoxia signaling cascade for erythropoietin production in hepatocytes.Mol Cell Biol 35:2658 –2672. doi:10.1128/MCB.00161-15.

Address correspondence to Norio Suzuki, sunorio@med.tohoku.ac.jp.

* Present address: Xiaoqing Pan, Division of Rare Cancer Research, National CancerCenter Research Institute, Tokyo, Japan; Tomokazu Souma, Division ofNephrology, Northwestern University Feinberg School of Medicine, Chicago,Illinois, USA; Shin-ichi Kawaguchi, Department of Applied Chemistry, GraduateSchool of Engineering, Osaka Prefecture University, Osaka, Japan.

Y.T. and H.S. contributed equally to this work.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

doi:10.1128/MCB.00161-15

2658 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

nomes contain three genes for HIF� isoforms. HIF1� and HIF2�have similar protein structures, and both exhibit strong transac-tivities under hypoxic conditions, whereas the function and regu-latory mechanisms of HIF3� remain controversial due to the ex-istence of multiple HIF3� splicing variants (18–20). Undernormal oxygenation conditions (normoxia), HIF� subunits areconsistently synthesized, and their specific proline residues arequickly hydroxylated by HIF-prolyl hydroxylase domain proteins(PHDs) (21, 22). Hydroxylated HIF� is recognized by von Hip-pel-Lindau protein (pVHL), which is a component of an E3 ubiq-uitin ligase complex, and degraded by the proteasome (23). Be-cause hypoxia inhibits the enzymatic activities of PHDs, which useoxygen as a substrate, HIF heterodimers have increased activitiesas transcription factors under hypoxic conditions (24).

Hepatic Epo production is enhanced by the experimental tar-geting of the HIF pathway using genetically modified mice, smallmolecules, or small interfering RNA s (siRNAs) (25). For instance,liver-specific pVHL deficiency in mice results in polycythemiawith the overexpression of the Epo gene in the liver (26). Becausethis phenotype depends on HIF2�, it is thought that HIF2�,rather than HIF1�, regulates hypoxia-inducible Epo expression inhepatocytes (26). The contributions of the three PHD isoforms(PHD1, PHD2 and PHD3, which are encoded by the Egln2, Egln1,and Egln3 genes, respectively) to hepatic Epo gene regulation havebeen investigated using gene-modified mice, and functional re-dundancy among the isoforms has been demonstrated (27–29).Although these strategies stimulate hepatic Epo production, theconstitutive activation of HIFs induces steatosis and ectopic ex-pression of cancer-related genes (25–29). Therefore, establishing aspecific strategy for hepatic Epo induction to treat renal anemia isimportant, and fully understanding the molecular mechanismunderlying Epo regulation will be necessary to develop such astrategy.

In this study, we analyzed gene-modified mouse lines lackingPHD isoforms, HIF2�, and/or EpoHE specifically in the liver to elu-cidate the in vivo signaling pathway between hypoxia sensor PHDsand cis-regulatory elements in the hepatic Epo gene regulatory sys-tem. Our results demonstrate that the function of PHDs is gene dosedependent and that HIF2� and EpoHE are necessary for hepatic Epoproduction. HIF2� functions in both the removal of histones fromthe region around the Epo gene, which enhances Epo transcription,and the induction of HIF3� expression to attenuate hepatic Epo pro-duction. Thus, we propose that a PHD-HIF2�-EpoHE axis regulateshepatic Epo production by controlling HIF3� expression and thechromatin structure of the Epo gene. Our results provide valuableinformation regarding therapeutic targets that enhance hepatic Epoproduction for renal-anemia patients.

MATERIALS AND METHODSMice. All mice were maintained at the Institute for Animal Experimenta-tion, Tohoku University Graduate School of Medicine, under the Regu-lations for Animal Experiments and Related Activities of Tohoku Univer-sity. A hypoxic box, in which the oxygen concentration was regulated byan oxygen controller (ProOx; BioSpherix) with a nitrogen generator(Nilox; Sanyo Electronic Industries), was used to expose the mice to hy-poxic conditions. For acclimation to severe hypoxia (6% O2), the micewere exposed to 10% oxygen for 2 h before the oxygen concentration wasreduced to 6%. Mouse lines carrying conditional knockout genes forPHDs (Egln1f/f, Egln2f/f, and Egln3f/f genotypes) and HIF2� (Epas1f/f ge-notype) were used (30, 31). The in vivo function of the Epo gene hepaticenhancer (EpoHE) was investigated using mice with a deletion of EpoHE

(�EpoHE mice, Epo�3=E/�3=E genotype) (9, 12). AlbCre transgenic micewere used to express Cre recombinase exclusively in mature hepatocytes(32). The Epas1f/f and AlbCre mice were purchased from the Jackson Lab-oratory. Genotyping of genomic DNA from tails was performed by PCRusing the primers listed in Table 1.

Reverse transcription and PCR. Total RNA samples from mouse or-gans were extracted using Isogen (Nippon Gene). cDNA was synthesizedusing a SuperScript III system (Invitrogen). To detect both normal andknockout transcripts derived from genes for PHDs simultaneously, cDNAswere amplified with primer pairs that annealed to sequences flanking thedeleted exons (Table 1), and the PCR products were subjected to gel elec-trophoresis, followed by ethidium bromide staining (30). PCR primersthat annealed to sequences in exons 3 and 4 of the Hif3a gene were used(Table 1) to detect both IPAS- and HIF3�-specific splicing variants of theHif3a gene by agarose gel electrophoresis because the IPAS mRNA con-tains 56-bp sequences from the IPAS-specific exon (exon 4a) that is be-tween exons 3 and 4 (33). Quantitative PCR (qPCR) was performed oncDNA with the primers listed in Table 1, using FastStart SYBR greenmaster mix with a LightCycler 96 system (Roche) or Power SYBR greenmaster mix with a StepOne real-time PCR system (Life Technologies).The expression levels of �-actin mRNA were used as internal controls forthe PCR experiments.

Blood analysis. Peripheral blood (0.2 to 0.3 ml) was collected from themouse submandibular vein into a 1.5-ml tube containing 5 �l of 0.5 MEDTA. After the blood samples were centrifuged, plasma Epo concentra-tions were measured using a mouse Epo enzyme-linked immunosorbentassay (ELISA) kit (R&D Systems). To induce anemia by bleeding, 0.3 ml ofperipheral blood was withdrawn from the mouse submandibular veinonce per day for 3 days, and then the recovery from the anemia wasinvestigated by measuring hematocrit levels in the retro-orbital venousplexus using heparinized microtubes (Drummond). The glucose concen-tration in blood from tail tips was directly measured using an animalglucometer (LAB Gluco; ForaCare).

Flow cytometry. Mononucleated cells from the spleen and bone mar-row were prepared by isopycnic centrifugation with Histopaque-1083(Sigma) and were stained with fluorescein isothiocyanate (FITC)-conju-gated Ter119 (116215; BioLegend) and phycoerythrin (PE)-conjugatedCD71 (113808; BioLegend) antibodies. After washing, the cells weresorted using a FACS Jazz flow cytometer (Becton Dickinson).

Immunoblotting. Nuclear extracts were obtained from mouse liversusing sucrose buffer (0.25 M sucrose in 10 mM HEPES-KOH buffer) andresuspended in radioimmunoprecipitation assay (RIPA) buffer (50 mMTris-Cl, 150 mM NaCl, 0.5% sodium deoxycholate, 0.1% SDS, and 1.0%NP-40) supplemented with inhibitors for proteinase (Roche) and protea-some (Proteintech). Whole-cell extracts were directly obtained with RIPAbuffer. The cell extracts were incubated with tobacco etch virus (TEV)protease (Promega) at 4°C overnight to remove the HaloTag fromHaloTag-HIF3� fusion proteins. Proteins (25 �g) were transferred tonitrocellulose membranes (Bio-Rad) after electrophoresis using 6% or8% polyacrylamide gels, and the membranes were incubated with primaryantibodies against HIF1� (PAB12138; Abnova), HIF2� (C150132;LSBio), HIF3� (HPA041141; Sigma), green fluorescent protein (GFP)(598; MBL), Nup62 (ab96134; Abcam), �-tubulin (T5168; Sigma) or�-tubulin (PA1-21153; Thermo). Horseradish peroxidase-conjugatedsecondary antibodies against mouse (01803-44; Nacalai Tesque), rabbit(P0448; Dako) or goat (sc-2020; Santa Cruz) immunoglobulins were usedwith chemiluminescent detection reagents (GE Healthcare) to detect sig-nals with a Chemi-Doc imaging system (Bio-Rad).

Glycogen assays. To measure the liver glycogen concentration, boiledsupernatants from liver homogenates (2.0 mg tissue/ml water) were di-gested with glucoamylase for 30 min at room temperature. Samples wereanalyzed using the fluorometric assay protocol in the glycogen assay kit(BioVision). To detect glycogen in liver sections, periodic acid-Schiff(PAS) staining was conducted on paraffin-embedded sections using a PASstaining kit (Muto).

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2659Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

Lipid droplet staining. Frozen sections (10-�m thickness) of mouselivers that had been fixed in 4% formaldehyde for 4 h were incubated withNile red solution (100 ng/ml; Wako) and Hoechst 33342 (1.0 �g/ml;Calbiochem) for 5 min (34). After the sections had been washed with PBS,fluorescence signals were detected using the pseudocolor detection systemof a BZ9000 microscope (Keyence).

Cell culture and reporter analysis. Hep3B cells that were derivedfrom a human hepatoma were cultured in high-glucose Dulbecco’s mod-ified Eagle medium (DMEM) (Nacalai Tesque) supplemented with 10%fetal bovine serum (Sigma) (35). For hypoxic cell sampling, the cells wereincubated and treated in a Sci-tive hypoxia workstation (Ruskinn), inwhich the air was composed of 1% O2, 5% CO2, and 94% N2.

The plasmid pT81/HRE-luc, which expresses firefly luciferase under thecontrol of HIF-binding sequences from EpoHE, was used as a reporter forEpoHE activity, and pCMX-HIF2�CA was used to overexpress Flag-taggedmouse HIF2� bearing constitutively activated mutations in proline residues(P405A and P530A) (36). We used the pFN21AB6219 plasmid (Kazusa Insti-tute) to express human HIF3� linked to a HaloTag peptide (Promega).

For transient-transfection assays, the cells were cultured to 70% con-fluence in 12-well plates and transfected with pT81/HRE-luc (300 ng/well), pCMX-HIF2�CA (300 ng/well), and/or pFN21AB6219 (300 or 600ng/well) using Lipofectamine 2000 (Life Technologies). For internaltransfection efficiency controls, 100 ng/well of the reference plasmidspEGFP-N1 (TaKaRa) and pRL-EF (37) were added to all transfectionmixtures. The cells were harvested at 48 h after transfection, and the ac-tivity of firefly luciferase was measured using a dual luciferase reporterassay system (Promega) after normalization with the activity of Renillaluciferase, which was expressed under the EEF1A1 gene promoter frompRL-EF (37). The cells were observed under a BZ9000 microscope aftersupplementation of the culture medium with tetramethylrhodamine(TMR)-conjugated HaloTag direct ligand (Promega) to detect HaloTag-HIF3� expression in living cells.

Chemical compounds. FG4592 (Selleck Chemicals) and acriflavinehydrochloride (Sigma) were dissolved in dimethyl sulfoxide (DMSO)(Nacalai Tesque). HIF2 antagonist 2 {N-(3-chloro-5-fluorophenyl)-4-ni-trobenzo[c]1,2,5-oxadiazol-5-amine; CAS no. 1422955-31-4} was synthe-

TABLE 1 Sequences of the primers used in this study

Purpose and target Primer 1 (5=–3=) Primer 2 (5=–3=)Genotyping

Egln1 (PHD2) CAAATGGAGATGGAAGATGC TCAACTCGAGCTGGAAACCEgln2 (PHD1) TGGGCGCTGCATCACCTGTATCT ACTGGTGAAGCCTGTAGCCTGTCEgln3 (PHD3) ATGGCCGCTGTATCACCTGTAT CCACGTTAACTCTAGAGCCACTGAAlbCre ACGTTCACCGGCATCAACGT CTGCATTACCGGTCGATGCAEpas1 (HIF2�) CAGGCAGCAGGCAGTATGCCTGGCTAATTCCAGTT CTTCTTCCATCATCTGGGATCTGGGACTEpoHE CAGGCTCCATTCAAGGC CCTGCAGTGGACTTTGAAGGC

RT-PCR with gel electrophoresisPHD1 ACCGCGCAGCATTCGTG GGGGCTGGCCATTAGGTAGGTGTAPHD2 GCGGGAAGCTGGGCAACTAC CAACCCTCACACCTTTCTCACCPHD3 CTGCGTGCTGGAGCGAGTCAA TCATGTGGATTCCTGCGGTCTGIPAS/HIF3� GAGGGTTTCGTCATGGTACT TCTTGAAGTTCCTCTTGGTC

RT-qPCR with SYBR green reagents for mouse(m) or human (h) mRNA expression

�-Actin (m, h) GACAGGATGCAGAAGGAGAT TTGCTGATCCACATCTGCTGPHD1 (m) ATGGCTCACGTGGACGCAGTAA CATTGCCTGGATAACACGCCACPHD2 (m) TAAACGGCCGAACGAAAGC GGGTTATCAACGTGACGGACAPHD3 (m) CTATGTCAAGGAGCGGTCCAA GTCCACATGGCGAACATAACCEpo (m) CATCTGCGACAGTCGAGTTCTG CACAACCCATCGTGACATTTTCEpoR (m) GGGCTCCGAAGAACTTCTGTG ATGACTTTCGTGACTCACCCTBnip3 (m) GTTACCCACGAACCCCACTTT GTGGACAGCAAGGCGAGAATVegfa (m) CTGCTGTAACGATGAAGCCCTG GCTGTAGGAAGCTCATCTCTCCFasn (m) GGCTGCTGTTGGAAGTCAGC AGTGTTCGTTCCTCGGAGTGScd1 (m) TTCTTGCGATACACTCTGGTGC CGGGATTGAATGTTCTTGTCGTHIF1� (m) CCTGCACTGAATCAAGAGGTTGC CCATCAGAAGGACTTGCTGGCTHIF2� (m) CAATGACAGCTGACAAGGAG CATAGAAGACCTCCGTCTCCHIF3� (m) AAGACGCCCTGACCCCCAGG CCCTCTGCTGGTGAGCGTGCGpi (m) CCCTCTTTATAATCGCCTCCA GAAACCACTCCTTTGCTGTCTCLdha (m) GGCACTGACGCAGACAAG TGATCACCTCGTAGGCACTGHepcidin (m) TCTTCTGCATTGGTATCGCA GAGCAGCACCACCTATCTCCCA9 (h) CCTTTGCCAGAGTTGACGAG GACAGCAACTGCTCATAGGCPGK1 (h) CTAACAAGCTGACGCTGGAC CTGGTTGTTTGTTATCTGGTTGEPO (h) TCATCTGTGACAGCCGAGTC CAAGCTGCAGTGTTCAGCAC

qPCR with SYBR green reagents for chromatinanalysis

CA9 promoter TCTGTGAGTCAGCCTGCTCC TCCCAGCACACGGTGTGTACPGK1 promoter CCCTAAGTCGGGAAGGTTCC CTGTCCGTCTGCGAGGGTACEPO promoter CTCAACCCAGGCGTCCTG GGGGCTGTTATCTGCATGTGEPO intron1 CTGTTTGAGCGGGGATTTAGC TTCCGGGGTCCTTGACAAGTEPO HE AACCTCCAAATCCCCTGGCTC CTGTGTGAGACAGCACGTAG

Tojo et al.

2660 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

sized from 3-chloro-5-fluoroaniline (Sigma-Aldrich) and 5-chloro-4-ni-trobenzo[c]1,2,5-oxadiazole (Namiki) in accordance with a previouslypublished protocol (38) and resuspended in DMSO after purification.

Chromatin analysis. Nuclear extracts of Hep3B cells were partiallydigested with nucleases from an EpiQ chromatin analysis kit (Bio-Rad) todetect the chromatin states of HIF target genes. Using purified genomicDNA from nuclease-digested and undigested samples, qPCR was con-ducted with the primers listed in Table 1, and the abundance of PCR targetsites within the two samples was compared. Because chromatin regionswith nucleosome structures (genomic DNA masked by histones) are re-sistant to nuclease digestion (accessibility), a similar number of target siteswere observed in the histone-rich regions in the nuclease-digested andundigested samples (100% protection from nuclease digestion). Nucleo-some-free chromatin regions show low levels of protection (39).

Statistics. The data are presented as means � standard deviations(SD). P values were calculated using two-tailed, unpaired Student’s t tests.

RESULTSPolycythemia is induced by the loss of two PHD isoforms in theliver. Mice harboring conditional knockout (floxed) genes forPHD1, PHD2, and PHD3 were crossed with AlbCre transgenicmice expressing Cre recombinase in hepatocytes to elucidate theroles of the PHD isoforms in hepatic Epo production (30, 32).Because loxP sequences were inserted at both ends of the exon that

encodes the catalytic domain for each PHD in the floxed alleles (f),the PHD transcripts that were produced following Cre-mediatedliver-specific knockout (LKO) were shorter than those generatedby the wild-type and floxed alleles (30). The shorter transcriptswere predominantly detected in the livers of triple-knockout mice(Egln2f/f:Egln3f/f:AlbCre� genotype; called P123-LKO mice),whereas the control mice, which did not carry the Cre transgene(Egln2f/f:Egln1f/f:Egln3f/f genotype), expressed only transcripts ofnormal length (Fig. 1A). These data indicated that the expressionof functional PHD isoforms was almost eliminated by AlbCretransgene expression in the hepatocytes of P123-LKO mice. Ad-ditionally, the levels of mRNA expression for each PHD isoform inthe livers of PHD mutant mice were quantitatively measured byreverse transcription-quantitative PCR (RT-qPCR), and the dataconfirmed the high efficiency of AlbCre-mediated liver-specificgene targeting (Fig. 1B). Interestingly, PHD2 mRNA expression inPHD1/PHD3 double-knockout (P13-LKO) livers was higher thanthat in the control mice. In addition, PHD3 mRNA expression wasincreased by the loss of the genes for the other PHD isoforms inlivers (Fig. 1B). These data suggest that the expression of the genesfor PHD2 and PHD3 is induced by HIFs (40, 41).

Next, we measured the hematocrit levels (red blood cell mass)

FIG 1 Liver-specific knockout of multiple PHD isoforms causes polycythemia. (A) RT-PCR was performed on livers from P123-LKO (Egln2f/f:Egln1f/f:Egln3f/f:AlbCre� genotype) or control (Cre� Egln2f/f:Egln1f/f:Egln3f/f genotype) mice. Then, the PCR products were subjected to agarose gel electrophoresis to detect theexpression of longer and shorter transcripts from the normal (w) and targeted (d [deletion of the exon encoding the catalytic domain of each PHD isoform])alleles, respectively. �-Actin was used as an internal control. (B) Average expression levels of PHD isoforms in the livers of the indicated genotype mice weremeasured by RT-qPCR. The values are averages for 3 to 5 mice in each group, with error bars indicating standard deviations (SD). *, P 0.01 compared to theexpression levels in the control mice, which are set at 100%. (C and D) Average hematocrit (C) and Epo concentration (D) in peripheral blood samples from PHDmutant mice at 8 to 19 weeks after birth. The values are averages for 4 to 12 mice in each group, with error bars indicating SD. *, P 0.01 compared to the levelsin the control mice.

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2661Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

in the peripheral blood samples of adult mice (11 to 39 weeks ofage) lacking a single PHD isoform in the liver (P1-LKO, P2-LKO,and P3-LKO mice). These mutant mice had no apparent abnor-malities, with normal hematocrit levels compared with controlmice lacking the Cre transgene (Fig. 1C). A double knockout ofPHD isoforms in any combination (P12-LKO, P13-LKO, andP23-LKO mice) resulted in polycythemia, with hematocrit levelsranging from 65% to 85% (Fig. 1C). Mice lacking all three PHDisoforms in the liver (P123-LKO mice) exhibited polycythemiawith a severity similar to that observed in the double-knockoutmice (Fig. 1C). These data demonstrate that the inhibition of twoPHD isoforms in the liver is necessary and sufficient for erythro-poietic induction and that the total dose of PHD isoforms in theliver seems to be more important for PHD suppressive activity inerythropoiesis than the specific function of each isoform.

We observed that the Epo concentration in the plasma of P123-LKO mice was significantly higher than that in control mice (Fig.1D). This result is consistent with the hematocrit data from poly-cythemic P123-LKO mice. However, plasma Epo levels in double-knockout mice were normal, although the mice suffered frompolycythemia (Fig. 1C and D).

Loss of two PHD isoforms induces Epo expression in theliver. Although the liver-specific double knockout of PHD iso-forms was thought to cause polycythemia through Epo oversecre-tion from the liver, the plasma Epo concentrations in these dou-ble-knockout mice were within the normal range (Fig. 1D). Thus,we measured the level of Epo expression in the livers of the PHDmutant mice. Epo expression was increased 10- to 20-fold by theloss of 2 PHD isoforms in the liver (Fig. 2A), while single PHDisoform knockout did not alter the Epo mRNA levels (28, 29). Inthe triple-knockout mice, the hepatic Epo mRNA level was dra-matically (approximately 200-fold) higher than levels in controland double-knockout mice (Fig. 2A). These data demonstratedthat PHD isoforms functionally compensate for each other in he-patic Epo production and that any single isoform can partiallysuppress hepatic Epo expression under normal conditions.

Next, we predicted that most of the circulating Epo in the dou-ble-knockout mice was consumed through the internalization ofEpo-EpoR complexes by an expanded population of erythroblas-tic cells in the spleen after erythropoiesis stimulation (42). As ex-pected, the spleens were significantly enlarged in the PHD double-knockout mice and in the triple-knockout mice (Fig. 2B and C).Splenic EpoR mRNA levels, which were normalized to �-actinmRNA levels, were extremely high in the double-knockout micecompared with control mice (Fig. 2D). EpoR expression was alsoslightly induced in the bone marrow of liver-specific PHD double-knockout mice (Fig. 2D). Flow cytometry of splenic mononu-cleated cells demonstrated that the Ter119� CD71� erythroblasticcell fraction of the double-knockout mice was approximately 10-fold higher than that in control mice (Fig. 2E) (13, 43). These datasuggest that the excess Epo produced by the PHD-deficient liversis absorbed by the expanded erythroblastic cells in spleens. Thishypothesis may explain why plasma Epo levels are normal in poly-cythemic mice lacking two PHD isoforms in their livers.

To test this hypothesis, the hematocrit levels in double-knock-out mice were reduced to normal by phlebotomy, and then thepostphlebotomy hematocrit levels were evaluated over time. Asexpected, the mice reverted to their original polycythemic pheno-type within 3 weeks postphlebotomy (Fig. 2F), and their plasmaEpo levels fell to within the normal range throughout the obser-

vation period. These findings suggest that these double-knockoutmice constitutively secrete excess Epo from the liver but that mostof the Epo is internalized by erythroblasts after initiating the sig-naling cascade that induces erythropoiesis. Because a large differ-ence in Epo mRNA levels between the double- and triple-knock-out livers exists (Fig. 2A), the rate of Epo production in P123-LKOlivers likely surpasses the rate of internalization.

Liver-specific loss of all PHD isoforms causes severe steatosisand growth retardation. Immunoblotting of liver nuclear ex-tracts demonstrated that protein levels of both HIF1� and HIF2�were increased by triple knockout of the PHD isoforms (Fig. 3A),while HIF proteins were undetectable in livers of the double-knockout, single-knockout, and control mice. These data are con-sistent with previous reports that HIF protein expression is verylow or undetectable in PHD double-knockout livers comparedwith triple-knockout livers (28, 29). Therefore, it is possible thatthe low-level accumulation of HIF1� and HIF2� is sufficient toinduce hepatic Epo gene expression and polycythemia in PHDdouble-knockout mice. Loss of all 3 PHD isoforms in hepatocytesresults in a drastic induction of Epo expression via the hyperacti-vation of HIFs. Thus, PHD isoforms compensate for each other inthe suppression of HIFs in hepatocytes under normal conditions,and no apparent abnormality is caused by loss of one of the threeisoforms. The presence of a single functional PHD isoform is in-sufficient for the oxygen-dependent inactivation of HIFs. HIF1�and/or HIF2� escapes degradation in the hepatocytes and, al-though present at undetectable levels, induces Epo overexpressionin the livers of double-knockout mice with polycythemia.

Because we showed that Epo expression was induced by HIFactivation in mice lacking more than two PHD isoforms in theliver, the expression of other HIF target genes was analyzed in thelivers of P23-LKO and P123-LKO mice. The results demonstratedthat the expression of Bnip3 and Vegfa mRNA was slightly in-creased in both P23-LKO and P123-LKO mouse livers comparedto control mouse livers (Fig. 3B). These data demonstrated thatPHD inactivation causes not only Epo production but also theglobal induction of HIF target genes, some of which promote mi-tochondrial autophagy (Bnip3) and angiogenesis (Vegfa).

The livers of P123-LKO mice at 7 weeks of age were large andfatty compared to the livers of mice in other groups. Nile redstaining of lipid droplets in the liver sections clearly showed thatP123-LKO mice suffered from severe steatosis (Fig. 3C). The dou-ble-knockout mice exhibited no lipid droplet accumulation in theliver (Fig. 3C). These data support previous reports showing thatthe liver-specific activation of the HIF pathway via the geneticinactivation of pVHL causes steatosis (44, 45). The expression ofthe genes for Fasn and Scd1, both of which are involved in fattyacid synthesis, was strongly suppressed in P123-LKO livers (Fig.3B). This finding suggests that mitochondrial function is inhibitedby HIF activation via Bnip3-mediated mitophagy (46) and thatthe accumulation of lipids due to decreased fatty acid usage inmitochondria may negatively regulate the expression of fatty acidsynthesis genes. Consistent with a previous report that demon-strated increased glycogen concentrations in response to pVHLdeficiency in liver (47), the loss of more than 2 PHD isoformsresulted in increased hepatic glycogen deposition (arrows in Fig. 3D).Indeed, the glycogen concentration in P23-LKO mouse livers wassignificantly higher than that in control mouse livers (Fig. 3E).

P123-LKO mice exhibited growth retardation (Fig. 3F) and ashortened life span of 14 weeks (Fig. 3G). These observations in-

Tojo et al.

2662 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

FIG 2 Overexpression of the Epo gene in livers lacking PHD isoforms. (A) Epo mRNA expression levels in the livers of PHD mutant mice at 8 to 12 weeks of age.The average expression level of control mice was set as 1.0 after normalization to �-actin mRNA levels. The data are means � SD (n 3 for each group). *, P 0.01 compared to the levels in the control mice. (B) Splenic hypertrophy in liver-specific PHD-deficient mice. Wet weights of spleens from the PHD mutant mice(7 to 9 weeks of age) were measured. The values are means and SD (n 3 for each group). *, P 0.01 compared to the weights of control mouse spleens. (C)Spleens from the PHD mutant mice were enlarged at 8 weeks of age. (D) EpoR mRNA levels in the spleens and bone marrow of control and P13-LKO mice. Theaverage expression level in control mouse spleens was set as 1.0 after normalization to �-actin mRNA levels. The data are means and SD (n 3 for each group).(E) Representative data from flow cytometry with CD71 and Ter119 antibodies in mononucleated cells from spleens of control and P13-LKO mice. Thepercentages of cells found within the circles are shown. (F) Changes in the hematocrit levels of the PHD mutant mice before (Pre) and after phlebotomy. Thehematocrit levels in mice at 12 weeks of age were decreased to approximately 30% by phlebotomy. Note that the hematocrit levels of PHD mutant mice returnedto polycythemic levels within 3 weeks after phlebotomy.

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2663Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

dicate that the universal knockout of all PHD isoforms in liversdramatically increases HIF activity, which causes severe steatosisand lethality. Because the inactivation of two PHD isoforms wassufficient to induce polycythemia without steatosis and lethality,we propose that mild inhibition of PHDs leads to the activation ofhepatic Epo production without severe side effects.

HIF2� is a major regulator of hypoxia-inducible Epo geneexpression in the liver. PHD mutant mice with the AlbCre trans-gene were mated with HIF2� conditional knockout mice to inves-tigate whether HIF1� or HIF2� activates the Epo gene in PHD-deficient hepatocytes (31). The polycythemia exhibited by the

double-knockout mice was completely rescued by the loss of theHIF2� gene (Epas1) in hepatocytes (P12H2-LKO, P13H2-LKO,and P23H2-LKO mice in Fig. 4A). The hematocrit levels ofHIF2�-deficient P123-LKO (P123H2-LKO) mice were lower thanthose of P123-LKO mice (Fig. 1C and 4A). Similar to the results ofdeleting the PHD isoforms in hepatocytes, pVHL deficiencycauses Epo overexpression and polycythemia in mice, and thisphenotype is blunted by the loss of HIF2� but not of HIF1� (26).Taken together, these data demonstrate that HIF2�, but notHIF1�, predominantly activates the hepatic Epo gene under hy-poxic conditions by signaling through PHDs. Because the lethality

FIG 3 Disruption of all three PHD isoforms in the liver causes steatosis and growth retardation. (A) Immunoblots for HIF1� and HIF2� in nuclear extracts fromthe livers of P123-LKO mice at the indicated ages. Nup62 was used as an internal control. (B) Expression levels of Bnip3, Vegfa, Fasn, and Scd1 mRNAs in thelivers of PHD mutant mice were analyzed by RT-qPCR. �-Actin was used as an internal control. The expression levels of control mice were set as 1.0. The dataare means and SD (n 3 for each group). (C) Lipid droplets in the livers of the PHD mutant mice at 7 weeks of age were stained with Nile red (red). Hoechst 33342was used for nuclear staining (blue). (D) Glycogen deposition in the livers of PHD mutant mice at 8 weeks of age was detected by PAS staining (arrows). Bars,500 �m (C) and 100 �m (D). (E) The glycogen concentration was measured in the livers of the control and P23-LKO mice at 8 weeks of age. The data are meansand SD (n 4 for each group). (F) Changes in body weight between 2 and 5 weeks after birth in P123-LKO and control mice. Error bars indicate SD (n 3 foreach group). (G) Kaplan-Meier survival analysis of the indicated genotype mice (n 20 for each group). More than half of the P123-LKO mice died by 10 weeksafter birth. *, P 0.01 compared with the control mice (B, E, and F).

Tojo et al.

2664 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

and severe steatosis in P123-LKO mice were also rescued byHIF2� deficiency, the hyperactivation of HIF2� most likely con-tributes to these two phenotypes.

Although the immunoblotting (Fig. 4B) and RT-qPCR (Fig.4C) data showed that HIF2� expression was almost absent inhepatocytes of P123H2-LKO mice, the overexpression of the Epogene and the polycythemia observed in P123-LKO mice were notcompletely rescued by HIF2� deletion (Fig. 4A and C). Thus, wesuggest that HIF1� accumulation may contribute to the overex-pression of Epo mRNA in the triple-knockout livers. The overex-pression of the Vegfa gene was also rescued by HIF2� deficiency inP123-LKO livers, while mRNA expression of the glycolytic en-zymes (Gpi1 and Ldha) was present at high levels regardless of theHIF2� status (Fig. 4C). These gene expression profiles indicatedthat HIF2�, rather than HIF1�, upregulates the transcription ofthe Epo and Vegfa genes in hepatocytes under hypoxic conditions,whereas the glycolytic genes are primarily regulated by HIF1�.Interestingly, the loss of all three PHD isoforms in hepatocytesstrongly (100-fold) increased the HIF3� mRNA level; this overex-pression was also rescued by HIF2� deficiency (Fig. 4C).

The Epo gene hepatic enhancer is required for Epo overpro-duction in PHD-deficient livers. We previously reported that anenhancer sequence that contains a HIF-binding sequence (hypox-ia-responsive element [HRE]) and that is proximally downstreamof the Epo gene transcriptional endpoint is necessary and suffi-

cient for the expression of the Epo gene in the liver (12). A mouseline (�EpoHE) lacking 500 bp of the hepatic enhancer for the Epogene (EpoHE) presents with fetal anemia due to a loss of Epoproduction by the fetal liver, which is the major Epo-producingsite in fetuses. In contrast, adult �EpoHE mice exhibit normalerythropoiesis because adult mice produce most of their Epo inthe kidneys, where Epo expression is independent of EpoHE (12).�EpoHE mice were crossed with liver-specific PHD-deficientmice to determine whether EpoHE is a cis-regulatory element forthe hepatic Epo expression induced by the PHD-HIF2� signalingcascade.

The polycythemia observed in PHD-mutant mice was com-pletely rescued by the loss of EpoHE (Fig. 5A), demonstrating theessential function of EpoHE in hepatic Epo induction by the PHD-HIF2� hypoxia signaling cascade. Indeed, Epo expression in thelivers of double (Fig. 5B)- and triple (Fig. 5C)-knockout mice withthe EpoHE deficiency fell within the normal range. These dataclearly showed that EpoHE is essential for the induction of hypox-ia-inducible Epo expression by the PHD-HIF2� signaling cascade.

HIF3� mRNA levels were unaffected by the loss of EpoHE (Fig.5C), indicating that EpoHE functions solely in Epo gene regula-tion. In addition, EpoHE loss did not rescue the severe steatosis(Fig. 5D), the elevated glycogen deposition (Fig. 5E), or the down-regulation of Scd1 expression (Fig. 5C) observed in PHD123-LKOlivers. Additionally, the lethality of P123-LKO mice was not res-

FIG 4 HIF2� is essential for the development of polycythemia in liver-specific PHD knockout mice. (A) Hematocrit levels in PHD and HIF2� mutant mice at8 to 16 weeks of age are shown as the means and SD (n 3 to 5 for each group). *, P 0.01 compared with control H2-LKO mice. (B) Immunoblots for HIF2�in nuclear extracts from the livers of PHD and HIF2� mutant mice at 4 to 10 weeks of age. Nup62 was used as an internal control. The data from 2 independentsamples are shown for each genotype group. (C) mRNA expression levels in the livers of PHD and HIF2� mutant mice were analyzed by RT-qPCR. �-Actin wasused as an internal control for RT-qPCR. The expression levels of control mice were set as 1.0. The data are means and SD (n 3 for each group). *, P 0.01compared with the control mice (white bar). Note that the overexpression of Epo mRNA in P123-LKO livers is attenuated by the loss of HIF2� (P123H2-LKO).

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2665Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

cued by the EpoHE deletion. These findings suggested that hepaticsteatosis and glycogen deposition are caused by the hyperactiva-tion of HIFs but are not related to Epo overproduction or to poly-cythemia.

The peripheral blood glucose concentration in the PHD dou-ble-knockout mice was decreased compared with that in the con-trol mice, and these reduced blood glucose levels were restored byEpoHE loss (Fig. 5F). This observation indicates that the decreasedblood glucose levels are caused by polycythemia but are not re-

lated to HIF hyperactivation in liver-specific PHD-deficient micebecause erythrocytes actively uptake and utilize blood glucose forenergy production (48).

Iron homeostasis is mainly regulated by hepcidin, which inhib-its iron usage for erythropoiesis, and hepcidin secretion from liv-ers is downregulated in mice actively undergoing erythropoiesis(49). Hepcidin mRNA levels were dramatically decreased in liversfrom polycythemic PHD double-knockout mice compared withcontrol mouse livers (Fig. 5G). In nonpolycythemic P23-LKO:

FIG 5 The hepatic enhancer for the Epo gene (EpoHE) is necessary for the development of polycythemia in PHD mutant mice. (A) Hematocrit levels in the PHDand EpoHE mutant mice at 6 to 15 weeks of age. The data are means � SD (n 3 for each group). (B) Epo mRNA expression levels in the livers of PHD and EpoHEmutant mice at 8 to 15 weeks of age. The expression levels of control mice were set as 1.0 after normalization to �-actin mRNA levels. The data are means � SD(n 3 for each group). *, P 0.01 compared with the levels in the control mice (left bar). (C) Expression levels of Epo, HIF3�, and Scd1 mRNAs in the liversof PHD and EpoHE mutant mice were analyzed by RT-qPCR. �-Actin was used as an internal control. The expression levels of control mice were set as 1.0. Thedata are means � SD (n 3 or 4 for each group). Note that the overexpression of Epo mRNA in P123-LKO livers is completely suppressed by the loss of EpoHE(P123-LKO:�EpoHE). (D) Lipid droplets in the livers of PHD mutant mice at 7 weeks after birth were stained with Nile red (red). Hoechst 33342 was used fornuclear staining (blue). (E) Glycogen deposition in the livers of the PHD and EpoHE mutant mice at 8 weeks of age was detected by PAS staining (purple). (F)Glucose concentrations were measured in peripheral blood from the PHD and EpoHE mutant mice at 8 weeks of age. The data are means � SD (n 3 for eachgroup). (G) Hepcidin mRNA expression in the livers of PHD and EpoHE mutant mice was analyzed by RT-qPCR. �-Actin was used as an internal control. Theexpression levels in the control mice were set as 1.0. The data are means � SD (n 3 or 4 for each group). Bars, 50 �m (D and E). *, P 0.01 compared with thecontrol mice (C, F, and G).

Tojo et al.

2666 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

�EpoHE mice, hepcidin mRNA levels were normal (Fig. 5G). Thisresult is consistent with a previous report that hepcidin expressionis suppressed by active erythropoiesis but not by direct effects ofEpo or HIFs (49).

HIF2� is involved in nucleosome reorganization around theEPO gene in hepatocytes. Next, we analyzed the chromatin stateof HIF target genes in human hepatoma Hep3B cells that expressboth HIF1� and HIF2� proteins in a hypoxia-inducible mannerto investigate how HIF2� activates transcription of the EPO genein hepatocytes (Fig. 6A). In Hep3B cells, the expression of EPO

and the expression of the well-known HIF targets carbonic anhy-drase 9 (CA9) and phosphoglycerate kinase 1 (PGK1) werestrongly induced by hypoxia (1% oxygen for 24 h) (Fig. 6B).HIF2�-specific inhibitor (HIF2 antagonist 2) (38) significantlydecreased CA9 and EPO mRNA levels under hypoxic conditions,while the PGK1 mRNA level was not affected (Fig. 6B). In con-trast, acriflavine, which is an inhibitor for both HIF1� and HIF2�(50), suppressed the expression of all three HIF target genes (Fig.6B). Because both compounds inhibit the heterodimerization ofHIF� and -� subunits (38, 50), these results corroborate the re-

FIG 6 A HIF2�-specific inhibitor blocks the dissociation of nucleosome structures in the EPO gene under hypoxic conditions in human hepatocytes. (A)Immunoblots for HIF1� and HIF2� in Hep3B human hepatoma cells under normoxic (N) and hypoxic (H24 [1% oxygen for 24 h]) conditions. �-Tubulin wasused as an internal control. Both HIF1� and HIF2� proteins accumulated in cells treated with a PHD inhibitor (FG4592) or in cells under hypoxic stimulation,and the protein levels were not altered by supplementation with HIF inhibitors (HIF2 antagonist 2 and acriflavine) under hypoxic conditions. (B) RT-qPCR ofCA9, PGK1, and EPO mRNA expression in Hep3B cells cultured with vehicle (v) or HIF inhibitors (HIF2 antagonist 2 and acriflavine) under normoxic (N) orhypoxic (H24) conditions. �-Actin was used as an internal control for RT-qPCR. The expression levels under normal conditions with vehicle supplementation(v/N) were set as 1.0. The data are means and SD from 4 independent experiments. *, P 0.01 compared with hypoxic samples supplemented with vehicle(v/H24). A HIF2�-specific inhibitor (HIF2 antagonist 2) strongly suppressed hypoxic induction of EPO expression but did not affect the expression of PGK1. (C)Nucleosome occupancy of chromatin loci for the CA9, PGK1, and EPO genes was detected via nuclease accessibility in Hep3B cells cultured with vehicle (v) orHIF inhibitors (HIF2 antagonist 2 and acriflavine) under normoxic (N) or hypoxic (H4 [1% oxygen for 4 h] and H24 [1% oxygen for 24 h]) conditions. Theindicated loci were detected by qPCR of nuclease-digested and undigested chromatin from Hep3B cells, and the amount found in undigested samples was set as100%. The values are means and SD from 4 independent experiments. *, P 0.01 compared with vehicle-treated samples (v) under each condition. Nucleaseaccessibility was high in the PGK1 promoter region in all samples, indicating that this region is free from nucleosome structure under both normoxic and hypoxicconditions. The promoter, gene body (intron 1), and hepatic enhancer (HE) of the EPO gene and the CA9 promoter region were well protected from nucleasedigestion under normal conditions (v/N), and nuclease accessibility increased under hypoxic conditions. Note that the formation of the nucleosome-free regionunder hypoxic conditions was repressed by HIF inhibitors.

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2667Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

sults from P123H2-LKO mice, which suggested that Epo gene ex-pression is primarily activated by HIF2� and not by HIF1�. In-deed, previous studies utilizing RNA interference demonstratedthat HIF2�-specific suppression, but not HIF1�-specific suppres-sion, dramatically reduces hypoxia-inducible Epo gene expressionin Epo-producing hepatic cell lines (51, 52).

Next, we used a qPCR-based technique to analyze the changesin nuclease accessibility (sensitivity) to chromatin DNA in Hep3Bcells that were exposed to hypoxia (39). Nuclease-sensitive sitesare often an indication of transcriptionally active genomic re-gions, in which histones have dissociated from nucleosomes toopen the chromatin for transcription factor binding. Under nor-mal conditions, chromatin DNA was highly protected from nu-clease digestion in the promoter, gene body (intron 1) and en-hancer (HE, a hepatic enhancer downstream of the transcriptionend site) regions of the EPO gene (Fig. 6C). Hypoxic stimuli (1%O2 for 4 or 24 h) increased the accessibility of the EPO gene locus(Fig. 6C). The CA9 promoter was similarly affected by hypoxia,whereas the PGK1 promoter remained sensitive to nucleases (Fig.6C). These data demonstrated that the CA9 and EPO genes inHep3B cells are covered by histones under normal conditions andthat the nucleosome structure is disassembled under hypoxic con-ditions, facilitating active transcription. The PGK1 promoter isalways nucleosome free to enable interactions with transcriptionfactors and to achieve the high level of gene expression.

HIF2 antagonist 2 blocked the hypoxia-induced formation ofopen chromatin structures in the EPO and CA9 gene loci (graybars in Fig. 6C). Acriflavine treatment also caused these chromatinregions to become insensitive to nuclease digestion under both 4-and 24-hour hypoxic conditions (Fig. 6C). These data corroborateour gene expression data showing that both inhibitors suppressEPO and CA9 gene expression. These inhibitors did not have anyeffect on the constitutively open chromatin locus of the PGK1gene (Fig. 6C). Thus, we concluded that HIF2� is required for theformation of open chromatin structures in the EPO gene locus toinduce active transcription under hypoxic conditions in hepato-cytes.

HIF3� is induced by hypoxia and attenuates HIF2�-medi-ated EPO transcription in hepatocytes. Because we discoveredthat HIF3� mRNA expression, similar to Epo mRNA expression,is strongly induced by the loss of PHD isoforms in livers in aHIF2�-dependent manner (Fig. 4C and 5C), the role of HIF3�function in hepatic Epo gene regulation was further analyzed. TheHif3a gene in mice mainly generates two splicing isoforms, IPASand NEPAS, in addition to full-length HIF3� (18, 19). NEPAS isexclusively expressed at embryonic and neonatal stages, whereasIPAS expression is induced by hypoxia in the hearts, lungs, andmuscles of adult mice (18). We determined the expression ofIPAS-specific transcripts in P123-LKO livers and found that IPASexpression was only weakly induced by PHD deficiency, com-pared to the strong induction of full-length HIF3� expressioncaused by PHD deficiency (Fig. 7A). HIF3� mRNA expression inthe mouse liver was also stimulated by exposure to hypoxic stress(6% oxygen for 48 h) (Fig. 7B). These data indicated that full-length HIF3� expression is induced by hypoxia through the PHD-HIF pathway in the liver.

We analyzed the activity of EpoHE in Hep3B cells that overex-press human full-length HIF3� to investigate the function ofHIF3� in EpoHE-mediated gene regulation. We detected the nu-cleus-specific localization of overexpressed HIF3�, which was

conjugated with HaloTag, via staining with a cell-permeableHaloTag ligand in living Hep3B cells (Fig. 7C). The expression ofa luciferase reporter driven by EpoHE was significantly increasedby HIF2� overexpression, and this induction was suppressed byHIF3� in a dose-dependent manner (Fig. 7D). These data indi-cated the functional importance of HIF3� in hypoxia-inducibleEpo gene regulation. After the removal of the HaloTag by TEVprotease, the Western blot data confirmed that the overexpressedHIF3� was the 80-kDa full-length HIF3� (HIF3�1) (20) and thatHIF2� protein levels in the constitutively active HIF2� mutantwere not affected by HIF3� overexpression (Fig. 7E). Taken to-gether, our results suggest that HIF3� expression in the liver isinduced by hypoxia to limit HIF2�-mediated Epo overproduc-tion (Fig. 7F).

DISCUSSION

The REP cells in the renal tubular interstitium are the site of themajority of the Epo production in adult animals (9, 13); the con-tribution of the liver, which is the other Epo-producing organ, toadult erythropoiesis is extremely small (12). Therefore, damage tothe REP cells in chronic kidney diseases causes renal anemia evenin patients with an intact liver. Strategies that inhibit PHDs byusing small molecules or RNA interference have been investigatedin an attempt to develop a treatment that would enhance hepaticEpo production in patients with renal anemia (14, 15, 53). How-ever, PHD inhibition induces several genes, including oncogenes,steatosis-related genes, and the Epo gene, via the activation of bothHIF1� and HIF2� (54). In the present study, we investigated thein vivo regulatory mechanisms of hepatic Epo production to iden-tify therapeutic targets that exclusively stimulate Epo production.

The liver-specific knockouts of each PHD isoform exhibited noapparent phenotype, indicating that PHDs have redundant func-tions in regulating the hypoxia-response pathway (28, 29). Theloss of all three genes for PHD isoforms in the liver resulted in Epooverproduction and polycythemia via the upregulation of bothHIF1� and HIF2�. However, the strong and constitutive activa-tion of HIFs caused steatosis of the liver and the expression ofother HIF target genes (29), and the mice died approximately 10weeks after birth. We speculate that the development of fatty liversin the P123-LKO mice is related to the HIF-mediated inactivationof mitochondria, which are the organelles that utilize intracellularfatty acids (55). We also showed that these phenotypes dependedon neither polycythemia nor Epo overproduction by examiningP123-LKO mice lacking EpoHE, which exhibited HIF overexpres-sion and lethal steatosis but not Epo overproduction. The datafrom the P123-LKO mice suggested that PHD inhibition is re-sponsible for the induction of hepatic Epo production, althoughuniversal PHD inhibitors not only induce Epo production butalso produce dangerous side effects.

Our data demonstrated that the disruption of two PHD iso-forms was sufficient to activate erythropoiesis via Epo gene induc-tion in hepatocytes, without the severe steatosis that was observedin the P123-LKO livers. Surprisingly, the mice lacking two PHDisoforms in any combination exhibited similar polycythemia phe-notypes, suggesting that no significant functional differences inhepatic Epo gene regulation exist among the three PHD isoforms.However, we believe that these isoforms may have specific roles inother parts of the hypoxia response pathway because these iso-forms have different affinities for HIFs, which differ in proteinstructure outside their catalytic domains (56).

Tojo et al.

2668 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

Although the level of Epo mRNA expression in the double-knockout livers was approximately 10-fold higher than that of thecontrol mice, their plasma Epo concentrations were within thenormal range. Epo is internalized and degraded in EpoR-express-ing erythroid cells after the transduction of erythropoietic signal-ing (42). Therefore, our data suggest that most of the Epo pro-

duced by the double-knockout livers is absorbed by erythroblasticcells expressing EpoR at a high level, which increase more than10-fold in number in the spleens of double-knockout mice. Theplasma concentration of thrombopoietin (Tpo) depends on thequantity of c-mpl (Tpo receptor)-expressing platelets in periph-eral blood (so-called “sponge theory”) (57). Thus, we propose a

FIG 7 Involvement of HIF3� in the hypoxic regulation of Epo expression. (A) Detection of the IPAS-specific splicing variant of the Hif3a gene by RT-PCR inlivers of P123-LKO and control mice. �-Actin was used as an internal control. The 207-bp PCR amplicon derived from the IPAS splicing variant was undetectedor detected at extremely low levels by agarose gel electrophoresis of PCR samples compared with the 151-bp amplicon derived from HIF3�-specific splicingvariants. (B) Expression levels of HIF3� mRNA in livers of wild-type mice exposed to normal (N) or hypoxic (6% oxygen) air for 48 h were measured byRT-qPCR. The expression levels of normoxic samples were set as 1.0. The data are means and SD from 3 independent mice for each condition. *, P 0.01compared with normoxic samples. (C) Overexpression of HIF3� in the nuclei of living Hep3B cells. Hep3B cells were transfected with a control plasmidexpressing GFP (pEGFP-N1 [green]) with (�) or without (�) the HaloTag-HIF3� expression plasmid. HaloTag-HIF3� was detected with a HaloTag-TMRdirect ligand (red) in Hoechst 33342-stained nuclei (blue [merged images in lower panels]). Bar, 10 �m. (D) Luciferase reporter analysis of EpoHE in Hep3B cellsoverexpressing HIF2� and/or HIF3�. The HIF2� expression plasmid and/or the HIF3� expression plasmid (�, 300 ng/well; ��, 600 ng/well) were cotrans-fected with the EpoHE-reporter plasmid. The relative luciferase activity of the control samples (white bar) was set as 1.0 after normalization with internal controls.The values are means and SD from 4 independent experiments. *, P 0.01 compared with control samples. (E) Expression of HIF2� and HIF3� in cells that wereused for the EpoHE reporter assay was confirmed by immunoblotting. The constitutively active HIF2� mutant was detected using anti-HIF2� antibody inwhole-cell lysates from the cells used for panel D. Dose-dependent expression of HaloTag-HIF3� (*) was detected using anti-HIF3� antibody. HaloTag-cleavedHIF3� (**) was also detected in a TEV protease-digested sample. GFP and �-tubulin were used as loading controls. (F) A model of hypoxia-inducible Epo generegulation in hepatocytes. Under normoxic conditions, PHDs suppress the activities of HIF1� and HIF2� via protein degradation, and nucleosome structureson the chromatin silence the Epo gene. In hypoxic cells, HIF1� and HIF2� are activated by the inactivation of PHDs. HIF2� induces Epo transcription via thedisassembly of the nucleosome structures. The HIF2�-mediated Epo induction is attenuated by HIF3�, which is also induced by HIF2�.

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2669Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

sponge theory for Epo, in which plasma Epo levels are related tothe number of erythroblastic cells in spleens.

Polycythemia in the liver-specific PHD knockout mice was res-cued by liver-specific HIF2� deficiency. This result is comparableto that of a previous study, which showed that HIF2� is requiredfor the development of polycythemia in liver-specific knockout ofpVHL, which is an E3 ligase of HIF� subunits (26). These findingsclearly indicate that HIF2�, rather than HIF1�, is the major trans-activator for Epo gene expression. Additionally, we showed thatpolycythemia was completely rescued by the loss of EpoHE. Takentogether, our data suggest that HIF2� binding to EpoHE, whichcontains an HRE, triggers Epo production in the liver and thatEpoHE is the only cis-regulatory element for hypoxia-inducibleEpo gene expression in hepatocytes (12). Patients with hepatocel-lular carcinoma occasionally present both polycythemia and Epooverproduction (58). In these cases, PHDs or other HIF suppres-sion systems might be disrupted in the malignant cells. Recently,hereditary polycythemia has been linked to mutations in thePHD2 and HIF2� genes (59–63), indicating that a PHD-HIF2�pathway regulates EPO gene expression in humans.

Chromatin analyses demonstrated that the genomic DNAaround the Epo gene, including the promoter and EpoHE, associ-ates with histones under normal conditions. Under hypoxic con-ditions, the nucleosome structure is disassembled in a HIF2�-dependent manner to activate Epo transcription (Fig. 7F).Chromatin remodeling factors, such as BRG1, BRM, and SWI/SNF, are implicated in HIF-mediated transcriptional activation(64, 65). Therefore, these factors may be recruited to the Epo genelocus by HIF2�, which binds to EpoHE in hypoxic cells.

We found that HIF3� expression is strongly induced by PHDdeficiency in the liver and that this induction requires HIF2�.Some reports have shown that HIF1�, but not HIF2�, activatesHif3a gene expression via binding to its HREs (20, 66). However,we suggest that hypoxia-inducible Hif3a gene expression in nor-mal mouse liver is regulated primarily by HIF2�. Because splicingvariants of HIF3� exist, the function of HIF3� in hypoxia-induc-ible gene regulation is controversial (18, 19, 67, 68). In the presentstudy, we demonstrated that full-length HIF3� is primarily in-duced in hypoxic livers and that this expression attenuates tran-scriptional activation driven by the HIF2�-EpoHE axis (Fig. 7F).Since HIF2� deletion did not completely reduce the overexpres-sion of the Epo gene, which was observed in P123-LKO mice, wesuggested that HIF1� secondly contributed to the overexpressionof Epo mRNA. Alternatively, loss of Hif3a gene induction byHIF2� deletion may be related to the incomplete reduction of theEpo overexpression in the quadruple-knockout livers.

As shown by the dramatic induction of hepatic Epo productionfollowing the genetic or pharmacological inactivation of PHDs,PHDs actively inhibit the function of HIFs, even in renal-anemiapatients. Therefore, it is plausible to inhibit PHDs in the liver totreat renal anemia. However, mouse genetic studies have demon-strated that the universal inhibition of PHDs strongly induces theactivation of HIFs and causes side effects, including steatosis (54,69). To avoid severe side effects caused by robust inhibition ofPHDs, one reasonable strategy suggested by this study is to estab-lish mechanisms to specifically induce HIF2�-EpoHE binding. Be-cause chemical inhibitors usually only partly suppress enzymaticactivity and their efficacy can be regulated by dosage, chemicalcompounds that inhibit PHDs have been tested in clinical trials forrenal anemia (14). Additionally, it was recently reported that

treating mice with a PHD inhibitor (FG4497) improves metabolicdysfunction, including hepatic steatosis (70). This evidence indi-cates that renal anemia can be treated with PHD inhibitors.

ACKNOWLEDGMENTS

We thank Atsuko Konuma, Mizuho Tanno, Eriko Naganuma, Aina Fu-kuda, and Koichiro Kato (Tohoku University). We are also grateful to theBiomedical Research Core and the Centre for Laboratory Animal Re-search of Tohoku University for technical support.

This work was supported in part by grants-in-aid from MEXT/JSPSKAKENHI (grants 26111002 and 24249015 to M.Y. and 26116702 and25670157 to N.S.), the Inamori Foundation (N.S.), the Takeda ScienceFoundation (N.S.), and the Platform for Drug Discovery, Informatics,and Structural Life Science from MEXT, Japan (T.T., T.M., M.Y., andN.S.).

The funders had no role in the design of the study, data collection andanalysis, decision to publish or preparation of the manuscript. We have nocompeting interests to declare.

Y.T. and N.S. designed the study. Y.T., H.S., I.H., X.P., T.S., N.T., andN.S. performed the experiments. K.T., G.-H.F., T.D., T.M., M.Y., and N.S.provided the gene-modified mouse lines. T.T, S.-I.K., and T.M. providedthe chemical compounds. Y.T., H.S., and N.S. analyzed the data and con-structed the figures. Y.T. and N.S. wrote the manuscript. M.I., T.M., M.Y.,and N.S. developed the project.

REFERENCES1. Ebert BL, Bunn HF. 1999. Regulation of the erythropoietin gene. Blood

94:1864 –1877.2. Milledge JS, Cotes PM. 1985. Serum erythropoietin in humans at high

altitude and its relation to plasma renin. J Appl Physiol 59:360 –364.3. Guidet B, Offenstadt G, Boffa G, Najman A, Baillou C, Hatzfeld C,

Amstutz P. 1987. Polycythemia in chronic obstructive pulmonary disease.Chest 92:867– 870. http://dx.doi.org/10.1378/chest.92.5.867.

4. Franke K, Gassmann M, Wielockx B. 2013. Erythrocytosis: the HIFpathway in control. Blood 122:1122–1128. http://dx.doi.org/10.1182/blood-2013-01-478065.

5. Wu H, Liu X, Jaenisch R, Lodish HF. 1995. Generation of committederythroid BFU-E and CFU-E progenitors does not require erythropoietinor the erythropoietin receptor. Cell 83:59 – 67.

6. Babitt JL, Lin HY. 2012. Mechanisms of anemia in CKD. J Am SocNephrol 23:1631–1634. http://dx.doi.org/10.1681/ASN.2011111078.

7. Suzuki N, Obara N, Yamamoto M. 2007. Use of gene-manipulated micein the study of erythropoietin gene expression. Methods Enzymol 435:157–177. http://dx.doi.org/10.1016/S0076-6879(07)35009-X.

8. Obara N, Suzuki N, Kim K, Nagasawa T, Imagawa S, Yamamoto M.2008. Repression via the GATA box is essential for tissue-specific erythro-poietin gene expression. Blood 111:5223–5232. http://dx.doi.org/10.1182/blood-2007-10-115857.

9. Pan X, Suzuki N, Hirano I, Yamazaki S, Minegishi N, Yamamoto M.2011. Isolation and characterization of renal erythropoietin-producingcells from genetically produced anemia mice. PLoS One 6:e25839. http://dx.doi.org/10.1371/journal.pone.0025839.

10. Asada N, Takase M, Nakamura J, Oguchi A, Asada M, Suzuki N,Yamamura K, Nagoshi N, Shibata S, Rao TN, Fehling HJ, Fukatsu A,Minegishi N, Kita T, Kimura T, Okano H, Yamamoto M, Yanagita M.2011. Dysfunction of fibroblasts of extrarenal origin underlies renal fibro-sis and renal anemia in mice. J Clin Invest 121:3981–3990. http://dx.doi.org/10.1172/JCI57301.

11. Souma T, Yamazaki S, Moriguchi T, Suzuki N, Hirano I, Pan X,Minegishi N, Abe M, Kiyomoto H, Ito S, Yamamoto M. 2013. Plasticityof renal erythropoietin-producing cells governs fibrosis. J Am Soc Neph-rol 24:1599 –1616. http://dx.doi.org/10.1681/ASN.2013010030.

12. Suzuki N, Obara N, Pan X, Watanabe M, Jishage K, Minegishi N,Yamamoto M. 2011. Specific contribution of the erythropoietin gene 3=enhancer to hepatic erythropoiesis after late embryonic stages. Mol CellBiol 31:3896 –3905. http://dx.doi.org/10.1128/MCB.05463-11.

13. Yamazaki S, Souma T, Hirano I, Pan X, Minegishi N, Suzuki N,Yamamoto M. 2013. A mouse model of adult-onset anaemia due to eryth-ropoietin deficiency. Nat Commun 4:1950. http://dx.doi.org/10.1038/ncomms2950.

Tojo et al.

2670 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

14. Bernhardt WM, Wiesener MS, Scigalla P, Chou J, Schmieder RE,Günzler V, Eckardt KU. 2010. Inhibition of prolyl hydroxylases increaseserythropoietin production in ESRD. J Am Soc Nephrol 21:2151–2156.http://dx.doi.org/10.1681/ASN.2010010116.

15. Querbes W, Rogorad RL, Moslehi J, Wong J, Chan AY, Bulgakova E,Kuchimanchi S, Akinc A, Fitzgerald K, Koteliansky Kaelin WG, Jr.2012. Treatment of erythropoietin deficiency in mice with systemicallyadministered siRNA. Blood 120:1916 –1922. http://dx.doi.org/10.1182/blood-2012-04-423715.

16. Semenza GL, Koury ST, Nejfelt MK, Gearhart JD, Antonarakis SE.1991. Cell-type-specific and hypoxia-inducible expression of the humanerythropoietin gene in transgenic mice. Proc Natl Acad Sci U S A 88:8725–8729. http://dx.doi.org/10.1073/pnas.88.19.8725.

17. Lendahl U, Lee KL, Yang H, Poellinger L. 2009. Generating specificityand diversity in the transcriptional response to hypoxia. Nat Rev Genet10:821– 832. http://dx.doi.org/10.1038/nrg2665.

18. Makino Y, Cao R, Svensson K, Bertilsson G, Asman M, Tanaka H, CaoY, Berkenstam A, Poellinger L. 2001. Inhibitory PAS domain protein is anegative regulator of hypoxia-inducible gene expression. Nature 414:550 –554. http://dx.doi.org/10.1038/35107085.

19. Yamashita T, Ohneda O, Nagano M, Iemitsu M, Makino Y, Tanaka H,Miyauchi T, Goto K, Ohneda K, Fujii KY, Poellinger L, Yamamoto M.2008. Abnormal heart development and lung remodeling in mice lackingthe hypoxia-inducible factor-related basic helix-loop-helix PAS proteinNEPAS. Mol Cell Biol 28:1285–1297. http://dx.doi.org/10.1128/MCB.01332-07.

20. Pasanen A, Heikkilä M, Rautavuoma K, Hirsilä M, Kivirikko KI,Myllyharju J. 2010. Hypoxia-inducible factor (HIF)-3alpha is subject toextensive alternative splicing in human tissues and cancer cells and isregulated by HIF-1 but not HIF-2. Int J Biochem Cell Biol 42:1189 –1200.http://dx.doi.org/10.1016/j.biocel.2010.04.008.

21. Epstein AC, Gleadle JM, McNeill LA, Hewitson KS, O’Rourke J, MoleDR, Mukherji M, Metzen E, Wilson MI, Dhanda A, Tian YM, MassonN, Hamilton DL, Jaakkola P, Barstead R, Hodgkin J, Maxwell PH, PughCW, Schofield CJ, Ratcliffe PJ. 2001. C. elegans EGL-9 and mammalianhomologs define a family of dioxygenases that regulate HIF by prolyl hy-droxylation. Cell 107:43–54.

22. Ivan M, Haberberger T, Gervasi DC, Michelson KS, Günzler V, Kondo K,Yang H, Sorokina I, Conaway RC, Conaway JW, Kaelin WG, Jr. 2002.Biochemical purification and pharmacological inhibition of a mammalianprolyl hydroxylase acting on hypoxia-inducible factor. Proc Natl Acad SciU S A 99:13459–13464. http://dx.doi.org/10.1073/pnas.192342099.

23. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell SJ, vonKriegsheim A, Hebestreit HF, Mukherji M, Schofield CJ, Maxwell PH,Pugh CW, Ratcliffe PJ. 2001. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation.Science 292:468 – 472.

24. Miyata T, Suzuki N, van Ypersele de Strihou C. 2013. Diabetic nephrop-athy: are there new and potentially promising therapies targeting oxygenbiology? Kidney Int 84:693–702. http://dx.doi.org/10.1038/ki.2013.74.

25. Haase HV. 2013. Regulation of erythropoiesis by hypoxia-inducible fac-tors. Blood Rev 27:47–53. http://dx.doi.org/10.1016/j.blre.2012.12.003.

26. Rankin EB, Biju MP, Liu Q, Unger TL, Rha J, Johnson RS, Simon MC,Keith B, Haase VH. 2007. Hypoxia-inducible factor-2 (HIF-2) regulateshepatic erythropoietin in vivo. J Clin Invest 117:1068 –1077. http://dx.doi.org/10.1172/JCI30117.

27. Takeda K, Aguila HL, Parikh NS, Li X, Lamothe K, Duan LJ, Takeda H,Lee FS, Fong GH. 2008. Regulation of adult erythropoiesis by prolylhydroxylase domain proteins. Blood 111:3229 –3235. http://dx.doi.org/10.1182/blood-2007-09-114561.

28. Minamishima YA, Kaelin WG, Jr. 2010. Reactivation of hepatic EPOsynthesis in mice after PHD loss. Science 329:407. http://dx.doi.org/10.1126/science.1192811.

29. Duan LJ, Takeda K, Fong GH. 2014. Hematological, hepatic, and retinalphenotypes in mice deficient for prolyl hydroxylase domain proteins inthe liver. Am J Pathol 184:1240 –1250. http://dx.doi.org/10.1016/j.ajpath.2013.12.014.

30. Takeda K, Ho VC, Takeda H, Duan LJ, Nagy A, Fong GH. 2006. Placentalbut not heart defects are associated with elevated hypoxia-inducible factor alevels in mice lacking prolyl hydroxylase domain protein 2. Mol Cell Biol26:8336–8346. http://dx.doi.org/10.1128/MCB.00425-06.

31. Gruber M, Hu CJ, Johnson RS, Brown EJ, Keith B, Simon MC. 2007.

Acute postnatal ablation of Hif-2alpha results in anemia. Proc Natl AcadSci U S A 104:2301–2306. http://dx.doi.org/10.1073/pnas.0608382104.

32. Postic C, Shiota M, Niswender KD, Jetton TL, Chen Y, Moates JM,Shelton KD, Lindner J, Cherrington AD, Magnuson MA. 1999. Dualroles for glucokinase in glucose homeostasis as determined by liver andpancreatic beta cell-specific gene knock-outs using Cre recombinase. JBiol Chem 274:305–315. http://dx.doi.org/10.1074/jbc.274.1.305.

33. Makino Y, Kanopka A, Wilson WJ, Tanaka H, Poellinger L. 2002.Inhibitory PAS domain protein (IPAS) is a hypoxia-inducible splicingvariant of the hypoxia-inducible factor-3alpha locus. J Biol Chem 277:32405–32408. http://dx.doi.org/10.1074/jbc.C200328200.

34. Greenspan P, Mayer EP, Fowler SD. 1985. Nile red: a selective fluores-cent stain for intracellular lipid droplets. J Cell Biol 100:965–973. http://dx.doi.org/10.1083/jcb.100.3.965.

35. Goldberg MA, Glass GA, Cunningham JM, Bunn HF. 1987. The regu-lated expression of erythropoietin by two human hepatoma cell lines. ProcNatl Acad Sci U S A 84:7972–7976.

36. Kallio PJ, Wilson WJ, O’Brien S, Makino Y, Poellinger L. 1999. Regu-lation of the hypoxia-inducible transcription factor 1alpha by the ubiqui-tin-proteasome pathway. J Biol Chem 274:6519 – 6525. http://dx.doi.org/10.1074/jbc.274.10.6519.

37. Ohtsuji M, Katsuoka F, Kobayashi A, Aburatani H, Hayes JD,Yamamoto M. 2008. Nrf1 and Nrf2 play distinct roles in activation ofantioxidant response element-dependent genes. J Biol Chem 283:33554 –33562. http://dx.doi.org/10.1074/jbc.M804597200.

38. Scheuermann TH, Li Q, Ma HW, Key J, Zhang L, Chen R, Garcia JA,Naidoo J, Longgood J, Frantz DE, Tambar UK, Gardner KH, Bruick RK.2013. Allosteric inhibition of hypoxia inducible factor-2 with small molecules.Nat Chem Biol 9:271–276. http://dx.doi.org/10.1038/nchembio.1185.

39. Chen P, Zhao J, Wang Y, Wang M, Long H, Liang D, Huang L, Wen Z,Li W, Li X, Feng H, Zhao H, Zhu P, Li M, Wang QF, Li G. 2013. H3.3actively marks enhancers and primes gene transcription via opening high-er-ordered chromatin. Genes Dev 27:2109 –2124. http://dx.doi.org/10.1101/gad.222174.113.

40. Metzen E, Stiehl DP, Doege K, Marxsen JH, Hellwig-Bürgel T, Jelk-mann W. 2005. Regulation of the prolyl hydroxylase domain protein 2(phd2/egln-1) gene: identification of a functional hypoxia-responsive el-ement. Biochem J 387:711–717. http://dx.doi.org/10.1042/BJ20041736.

41. Pescador N, Cuevas Y, Naranjo S, Alcaide M, Villar D, Landázuri MO, DelPeso L. 2005. Identification of a functional hypoxia-responsive element thatregulates the expression of the egl nine homologue 3 (egln3/phd2) gene.Biochem J 390:189–197. http://dx.doi.org/10.1042/BJ20042121.

42. Gross AW, Lodish HF. 2006. Cellular trafficking and degradation oferythropoietin and novel erythropoiesis stimulating protein (NESP). JBiol Chem 281:2024 –2032. http://dx.doi.org/10.1074/jbc.M510493200.

43. Suzuki N, Suwabe N, Ohneda O, Obara N, Imagawa S, Pan X, Moto-hashi H, Yamamoto M. 2003. Identification and characterization of 2types of erythroid progenitors that express GATA-1 at distinct levels.Blood 102:3575–3583. http://dx.doi.org/10.1182/blood-2003-04-1154.

44. Haase VH, Glickman JN, Socolovsky M, Jaenisch R. 2001. Vasculartumors in livers with targeted inactivation of the von Hippel-Lindau tu-mor suppressor. Proc Natl Acad Sci U S A 98:1583–1588.

45. Qu A, Taylor M, Xue X, Matsubara T, Metzger D, Chambon P,Gonzalez FJ, Shah YM. 2011. Hypoxia-inducible transcription factor 2�promotes steatohepatitis through augmenting lipid accumulation, in-flammation, and fibrosis. Hepatology 54:472– 483. http://dx.doi.org/10.1002/hep.24400.

46. Zhang J, Ney PA. 2009. Role of BNIP3 and NIX in cell death, autophagy,and mitophagy. Cell Death Differ 16:939 –946. http://dx.doi.org/10.1038/cdd.2009.

47. Park SK, Haase VH, Johnson RS. 2007. von Hippel Lindau tumor sup-pressor regulates hepatic glucose metabolism by controlling expression ofglucose transporter 2 and glucose 6-phosphatase. Int J Oncol 30:341–348.http://dx.doi.org/10.3892/ijo.30.2.341.

48. Zhang JZ, Ismail-Beigi F. 1998. Activation of Glut1 glucose transporter inhuman erythrocytes. Arch Biochem Biophys 356:86 –92. http://dx.doi.org/10.1006/abbi.1998.0760.

49. Liu Q, Davidoff O, Niss K, Haase VH. 2012. Hypoxia-inducible factorregulates hepcidin via erythropoietin-induced erythropoiesis. J Clin In-vest 122:4635– 4644. http://dx.doi.org/10.1172/JCI63924.

50. Lee K, Zhang H, Qian DZ, Rey S, Liu JO, Semenza GL. 2009. Acriflavineinhibits HIF-1 dimerization, tumor growth, and vascularization. Proc

Hypoxia Signaling for Hepatic Epo Production

August 2015 Volume 35 Number 15 mcb.asm.org 2671Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from

Natl Acad Sci U S A 106:17910 –17915. http://dx.doi.org/10.1073/pnas.0909353106.

51. Befani C, Mylonis I, Gkotinakou IM, Georgoulias P, Hu CJ, Simos G,Liakos P. 2013. Cobalt stimulates HIF-1-dependent but inhibits HIF-2-dependent gene expression in liver cancer cells. Int J Biochem Cell Biol45:2359 –2368. http://dx.doi.org/10.1016/j.biocel.2013.07.025.

52. Warnecke C, Zaborowska Z, Kurreck J, Erdmann VA, Frei U, WiesenerM, Eckardt KU. 2004. Differentiating the functional role of hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha (EPAS-1) by the use ofRNA interference: erythropoietin is a HIF-2alpha target gene in Hep3Band Kelly cells. FASEB J 18:1462–1464. http://dx.doi.org/10.1096/fj.04-1640fje.

53. Vachal P, Miao S, Pierce JM, Guiadeen D, Colandrea VJ, Wyvratt MJ,Salowe SP, Sonatore LM, Milligan JA, Hajdu R, Gollapudi A, KeohaneCA, Lingham RB, Mandala SM, DeMartino JA, Tong X, Wolff M,Steinhuebel D, Kieczykowski GR, Fleitz FJ, Chapman K, Athanasopou-los J, Adam G, Akyuz CD, Jena DK, Lusen JW, Meng J, Stein BD, XiaL, Sherer EC, Hale JJ. 2012. 1,3,8-Triazaspiro[4.5]decane-2,4-diones asefficacious pan-inhibitors of hypoxia-inducible factor prolyl hydroxylase1-3 (HIF PHD1-3) for the treatment of anemia. J Med Chem 55:2945–2959. http://dx.doi.org/10.1021/jm201542d.

54. Rankin EB, Rha J, Selak MA, Unger TL, Keith B, Liu Q, Haase VH.2009. Hypoxia-inducible factor 2 regulates hepatic lipid metabolism. MolCell Biol 29:4527– 4538. http://dx.doi.org/10.1128/MCB.00200-09.

55. Semenza GL. 2011. Regulation of metabolism by hypoxia-inducible fac-tor 1. Cold Spring Harbor Symp Quant Biol 76:347–353. http://dx.doi.org/10.1101/sqb.2011.76.010678.

56. Pappalardi MB, Mcnulty DE, Martin JD, Fisher KE, Jiang Y, Burns MC,Zhao H, Thau HO, Sweitzer S, Schwartz Annan B RS, Copeland RA,Tummino PJ, Luo L. 2011. Biochemical characterization of human HIFhydroxylases using HIF protein substrates that contain all three hydroxylationsites. Biochem J 436:363–369. http://dx.doi.org/10.1042/BJ20101201.

57. Fielder PJ, Gurney AL, Stefanich E, Marian M, Moore MW, Carver-Moore K, de Sauvage FJ. 1996. Regulation of thrombopoietin levels byc-mpl-mediated binding to platelets. Blood 87:2154 –2161.

58. Sakisaka S, Watanabe M, Tateishi H, Harada M, Shakado S, Mimura Y,Gondo K, Yoshitake M, Noguchi K, Hino T. 1993. Erythropoietinproduction in hepatocellular carcinoma cells associated with polycythe-mia: immunohistochemical evidence. Hepatology 18:1357–1362.

59. Percy MJ, Zhao Q, Flores A, Harrison C, Lappin TR, Maxwell PH,McMullin MF, Lee FS. 2006. A family with erythrocytosis establishesa role for prolyl hydroxylase domain protein 2 in oxygen homeostasis.Proc Natl Acad Sci U S A 103:654 – 659. http://dx.doi.org/10.1073/pnas.0508423103.

60. Ladroue C, Hoogewijs D, Gad S, Carcenac R, Storti F, Barrois M,Gimenez-Rogueplo AP, Leporrier M, Casadevall N, Hermine O. 2012.Distinct deregulation of the hypoxia inducible factor by PHD2 mutantsidentified in germline DNA of patients with polycythemia. Haematologi-caae 97:9 –14. http://dx.doi.org/10.3324/haematol.2011.044644.

61. Albiero E, Ruggeri M, Fortuna S, Finotto S, Bernardi M, Madeo D,Rodeghiero F. 2012. Isolated erythrocytosis: study of 67 patients andidentification of three novel germ-line mutations in the prolyl hydroxylasedomain protein 2 (PHD2) gene. Haematologicaae 97:123–127. http://dx.doi.org/10.3324/haematol.2010.039545.

62. Percy MJ, Chung YJ, Harrison C, Mercieca J, Hoffbrand AV, DinardoCL, Santos PC, Fonseca GH, Gualandro SF, Pereira AC, Lappin TR,McMullin MF, Lee FS. 2012. Two new mutations in the HIF2A geneassociated with erythrocytosis. Am J Hematol 87:439 – 442. http://dx.doi.org/10.1002/ajh.23123.

63. Perrotta S, Stiehl DP, Punzo F, Scianguetta S, Borriello A, BencivengaD, Casale M, Nobili B, Fasoli S, Balduzzi A. 2013. Congenital erythro-cytosis associated with gain-of-function HIF2A gene mutations and eryth-ropoietin levels in the normal range. Haematologicaae 98:1624 –1632.http://dx.doi.org/10.3324/haematol.2013.088369.

64. Wang F, Zhang R, Beischlag TV, Muchardt C, Yaniv M, Hankinson O.2004. Roles of Brahma and Brahma/SWI2-related gene 1 in hypoxic in-duction of the erythropoietin gene. J Biol Chem 279:46733– 46741. http://dx.doi.org/10.1074/jbc.M409002200.

65. Sena JA, Wang L, Hu CJ. 2013. BRG1 and BRM chromatin-remodelingcomplexes regulate the hypoxia response by acting as coactivators for asubset of hypoxia-inducible transcription factor target genes. Mol CellBiol 33:3849 –3863. http://dx.doi.org/10.1128/MCB.00731-13.

66. Makino Y, Uenishi R, Okamoto K, Isoe T, Hosono O, Tanaka H,Kanopka A, Poellinger L, Haneda M, Morimoto C. 2007. Transcrip-tional up-regulation of inhibitory PAS domain protein gene expression byhypoxia-inducible factor 1 (HIF-1): a negative feedback regulatory circuitin HIF-1-mediated signaling in hypoxic cells. J Biol Chem 282:14073–14082. http://dx.doi.org/10.1074/jbc.M700732200.

67. Tanaka T, Wiesener M, Bernhardt W, Eckardt KU, Warnecke C. 2009.The human HIF (hypoxia-inducible factor)-3alpha gene is a HIF-1 targetgene and may modulate hypoxic gene induction. Biochem J 424:143–151.http://dx.doi.org/10.1042/BJ20090120.

68. Maynard MA, Qi H, Chung J, Lee EH, Kondo Y, Hara S, Conaway RC,Conaway JW, Ohh M. 2003. Multiple splice variants of the human HIF-3alpha locus are targets of the von Hippel-Lindau E3 ubiquitin ligase complex.J Biol Chem 278:11032–11040. http://dx.doi.org/10.1074/jbc.M208681200.

69. Minamishima YA, Moslehi J, Padera RF, Bronson RT, Liao R, KaelinWG, Jr. 2009. A feedback loop involving the Phd2 prolyl hydroxylasetunes the mammalian hypoxic response in vivo. Mol Cell Biol 29:5729 –5741. http://dx.doi.org/10.1128/MCB.00331-09.

70. Rahtu-Korpela L, Karsikas S, Hörkkö S, Blanco Sequeiros R, Lammen-tausta E, Mäkelä KA, Herzig KH, Walkinshaw G, Kivirikko KI, Mylly-harju J, Serpi R, Koivunen P. 2014. HIF prolyl 4-hydroxylase-2 inhibi-tion improves glucose and lipid metabolism and protects against obesityand metabolic dysfunction. Diabetes 63:3324 –3333. http://dx.doi.org/10.2337/db14-0472.

Tojo et al.

2672 mcb.asm.org August 2015 Volume 35 Number 15Molecular and Cellular Biology

on January 31, 2018 by guesthttp://m

cb.asm.org/

Dow

nloaded from